We propose to evaluate state-of-the-art NMR imaging for breast cancer detection. Low dose-film/screen mammography and NMR imaging will be compared in a clinical setting to determine the optimal role of NMR. We are particularly interested in the usefulness of NMR in evaluating cases of indeterminate mammography. The work will be accomplished at Washington University School of Medicine. Approximately 200 women will receive NMR evaluations over a two year period. Women will be randomly selected from 9 strate for participation. The sampling population will consist of approximately 3000 woman referred to our Mallinckrodt Institute of Radiology for mammography during the first 18 months of the project. All women referred will receive breast palpation and mammorgraphy. Informed consent will be required for NMR evaluation. Siemens Mammomat (mammography) and Siemens Magnetom NMR systems at Mallinckrodt will be used. For evaluation purposes images will be interpreted independently for each modality by different radiologists. Radiologists and pathologists will collaborate in the tissue classification efforts. Collaboration with NASA Kennedy Space Center image processing engineers will result in modifications and transfer of exceptional spectral analysis (tissue classification) software to our NMR imager early in the project. The new software will facilitate interpretation and increase NMR sensitivity and specificity. Pilot studies with NASA have already begun.